基于UMAP辅助的模糊C聚类方法进行太赫兹光谱识别

太赫兹(THz)具有低能性、瞬态性、波谱分析能力强的优点,在物质鉴别方面具有广阔的应用前景。现有的基于THz的物质鉴别方法,虽然取得了一定的效果,但是存在容易陷入局部最优的问题,从而导致识别精度不高。均匀流形逼近与投影(UMAP)作为一种非线性降维方法,其假设数据均匀分布在黎曼流形上,可以对具有模糊拓扑结构的流形进行建模。UMAP降维的过程是通过最小化两个拓扑表示之间的交叉熵,从而实现低维空间中数据表示的布局优化。传统的模糊C聚类方法(FCM)在聚类时,初始聚类中心往往随机给定,当初始聚类中心选择不恰当时,容易导致错误的聚类。为此,提出一种基于UMAP辅助的模糊C聚类算法,首先运用UMAP对输入的THz样本矩阵进行降维;再根据类与类之间距离最大化的原则,选择合适的初始聚类中心;最后利用模糊C均值聚类的方法进行聚类。所提出的方法不仅能够解决聚类过程中类与类之间过度拥挤的现象,而且能够反映出类别间的距离信息以便于给样本选择合适的初始聚类中心。为了验证提出的聚类方法的可靠性,运用太赫兹时域光谱技术对鲁棉研28、鲁棉研29、鲁棉研36、中棉28四种不同类型的转基因棉花种子进行了探测,利用基于U...     

基于SERS技术的食源性致病菌芽孢拉曼光谱特征结构分析及快速识别

为了探究食源性致病菌芽孢的拉曼特征指纹图谱,实现快速识别,该研究以产气荚膜梭菌(C.perfringens)、艰难梭菌(C.difficile)和蜡样芽孢杆菌(B.cereus)的芽孢为研究对象,以柠檬酸钠还原法制备的AgNPs溶胶为基底材料,用SERS技术对芽孢进行拉曼光谱检测,解析食源性致病菌芽孢的分子结构、不同芽孢之间的异同之处。将3种食源性致病菌芽孢的SERS光谱与主成分分析(PCA)和系统聚类分析(HCA)相结合并进行对比分析,实现不同种属食源性致病菌芽孢的定性识别。结果表明,不同食源性致病菌芽孢的SERS光谱的特异性和重现性良好。芽孢光谱中Ca²⁺-DPA的拉曼振动峰数量和峰强度占主要地位,其拉曼振动峰位置在657～663, 818～820, 1 017, 1 389～1 393, 1 441～1 449和1 572～1 576 cm^-1波段。C.difficile spores SERS光谱中Ca²⁺-DPA的六个特征峰峰强度均高于C.perfringens spores和B.cereus spores,C...     

Identification and control of nonlinear systems by a dissimilation particle swarm optimization-based Elman neural network

In this paper, we first present a learning algorithm for dynamic recurrent Elman neural networks based on a dissimilation particle swarm optimization. The proposed algorithm computes concurrently both the evolution of network structure, weights, initial inputs of the context units, and self-feedback coefficient of the modified Elman network. Thereafter, we introduce and discuss a novel control method based on the proposed algorithm. More specifically, a dynamic identifier is constructed to perform speed identification and a controller is designed to perform speed control for Ultrasonic Motors (USM). Numerical experiments show that the novel identifier and controller based on the proposed algorithm can both achieve higher convergence precision and speed than other state-of-the-art algorithms. In particular, our experiments show that the identifier can approximate the USM's nonlinear input–output mapping accurately. The effectiveness of the controller is verified using different kinds of speeds of constant, step, and sinusoidal types. Besides, a preliminary examination on a randomly perturbation also shows the robust characteristics of the two proposed models.     

Logistic和Lotka-Volterra模型参数的灰色估计方法研究

根据灰色系统建模原理,对Logistic模型中的参数进行了估计,将此方法应用于Gause(1934)草履虫实验数据的拟合,并与已有拟合结果对比,说明此方法是有效的.然后将其推广至两种群的Lotka—Volterra模型参数的确定,来研究山西省人口—经济系统的发展演化状况,得出了满意的结果.     

基于深度学习的太赫兹时域光谱识别研究

太赫兹时域光谱技术,由于其具有物质"指纹谱"特性,是一种可以快速无损地鉴别物质的重要手段,在毒品和爆炸物的无损检测等方面有广阔的应用前景.其中,光谱识别是太赫兹时域光谱技术应用研究的重要方向之一.现有的光谱识别方法多是依靠手工选取特征后进行机器学习分类,或是通过设置吸收峰阈值门限进行判断.由于一些物质在太赫兹波段内并没...     

EMD与神经网络在气液两相流流型识别中的应用

本文提出了EMD与Elman神经网络相结合的气液两相流流型识别的新方法.将压差波动信号经验模态分解(EMD)后的固有模态函数(IMF)进行分析、提取IMF能量作为Elman神经网络的输入特征向量,对水平管内的气液两相流流型进行识别.实验结果表明:该方法优于BP网络且稳定、识别率高,具有可行性.     

Encapsulation of nanocrystals with responsive gels for spatial optical identification

A tripeptide (β-AspFF) gelator was designed and synthesised using solid-phase peptide synthesis. The as-prepared gelator formed a stable and clear multi-responsive gel in toluene solution. Encapsulating the quantum dots with β-AspFF effectively increased stability and antioxidisation ability which are beneficial for optical identification.     

A data-driven,high-throughput methodology to determine tissue-specific differentially methylated regions able to discriminate body fluids

Tissue-specific differentially methylated regions (tDMRs) are regions of the genome with methylation patterns that modulate gene expression in those tissue types. The detection of tDMRs in forensic evidence can permit the identification of body fluids at trace levels. In this report, we have performed a bioinformatic analysis of an existing array dataset to determine if new tDMRs could be identified for use in body fluid identification from forensic evidence. Once these sites were identified, primers were designed and bisulfite modification was performed. The relative methylation level for each body fluid at a given locus was then determined using qPCR with high-resolution melt analysis (HRM). After screening 127 tDMR's in multiple body fluids, we were able to identify four new markers able to discriminate blood (2 markers), vaginal epithelia (1 marker) and buccal cells (1 marker). One marker for each target body fluid was also tested with pyrosequencing showing results consistent with those obtained by HRM. This work successfully demonstrates the ability of in silico analysis to develop a novel set of tDMRs capable of being differentiated by real time PCR/HRM. The method can rapidly determine the body fluids left at crime scenes, assisting the triers of fact in forensic casework.     

Breaking chaotic shift key communication via adaptive key identification

This paper proposes an adaptive parameter identification method for breaking chaotic shift key communication from the transmitted signal in public channel. The sensitive dependence property of chaos on parameter mismatch is used for chaos adaptive synchronization and parameter identification. An index function about the synchronization error is defined and conjugate gradient method is used to minimize the index function and to search the transmitter's parameter (key). By using proposed method, secure key is recovered from transmitted signal generated by low dimensional chaos and hyper chaos switching communication. Multi-parameters can also be identified from the transmitted signal with noise.     

Perspective methods of human identification: Ear biometrics

Geometrical methods of feature extraction from ear images in order to perform human identification are presented. Geometrical approach is motivated by the actual procedures used by police and forensic experts (so-called ear otoscopy). In their work, geometrical features of ears such as size, height, width, and shapes of earlobe are useful and valid proofs of identity. The contribution of the article is development of the new and original methods of geometrical feature extraction from 2D ear images. Four novel algorithms of ear feature extraction from contour images are described in detail. Moreover, identification results obtained for each of the methods, based on the distance of feature vectors in the feature space, are presented.